Data-driven intelligent evaluation method and system for geotechnical slope reinforcement quality

By calculating the similarity degree and maximum point distance between the ultrasonic transmitted signal and the echo signal, and adjusting the clustering measurement distance in the geographical location, the problem of clustering results deviation in the evaluation of geotechnical slope reinforcement is solved, and the evaluation accuracy is improved sex.

CN118586784BInactive Publication Date: 2024-12-06JINAN RAILWAY TRANSPORT GRP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411067060.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art fails to fully consider the change relationship and geographical connection between the echo signals of the places to be measured in the evaluation of geotechnical slope reinforcement quality, resulting in deviations in clustering results and reducing the accuracy of the evaluation.

Method used

By obtaining the similarity degree and maximum point distance between the ultrasonic transmitted signal and the echo signal, calculating the abnormal characteristic value, and adjusting the clustering metric distance in combination with geographical location, signal clustering is performed to improve evaluation accuracy.

Benefits of technology

The accuracy of the quality evaluation of rock and soil slope reinforcement is improved, clustering results errors caused by geographical location are avoided, and a more accurate quality evaluation of reinforcement is obtained.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118586784B_ABST
    Figure CN118586784B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of electromagnetic signal processing technology, and in particular to a data-driven intelligent evaluation method and system for the reinforcement quality of rock and soil slopes. The method obtains the abnormal characteristic value of the ultrasonic echo signal at the location to be tested, and obtains the initial clustering metric distance according to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal characteristic values; the initial clustering metric distance is adjusted according to the distance between the geographical locations corresponding to any two ultrasonic echo signals to obtain the corrected clustering metric distance; the ultrasonic echo signals are clustered based on the corrected clustering metric distance to obtain signal clustering clusters; and the reinforcement quality of the rock and soil slope is evaluated based on the comprehensive abnormality of the obtained signal clustering clusters. The present invention takes into account the spatial distance between ultrasonic echo signals, improves the accuracy of the clustering results based on the corrected clustering metric distance, and improves the accuracy of the reinforcement quality evaluation of the rock and soil slope.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of electromagnetic signal processing, and in particular to a data-driven intelligent evaluation method and system for rock and soil slope reinforcement quality. Background Art

[0002] Geotechnical slopes refer to natural mountains or artificially excavated soil or rock slopes. Under the influence of gravity and other external factors, they may become loose, collapse, or deform, which may lead to engineering accidents. Therefore, it is necessary to analyze the reinforcement quality of geotechnical slopes to prevent safety hazards.

[0003] In the existing method, an ultrasonic sensor installed inside a drone is used to transmit an ultrasonic signal to the test site at a certain height above the test site of the rock and soil slope. The ultrasonic wave is reflected by the test site of the rock and soil slope to generate an echo. The detector carried by the drone measures the echo signal absorbed and reflected by the test site to obtain relevant information about the reinforcement of the rock and soil slope. The characteristic value of the echo signal is obtained by analyzing the difference between the ultrasonic signal and the echo signal at the test site. The existing method usually directly clusters the echo signal using the characteristic value. The amplitude of the data point of the echo signal shows the change law of the echo signal, and the geographical environment and soil moisture of the test sites with similar locations are similar, that is, the slope reinforcement quality of the test sites with similar locations is similar. This feature reflects the geographical connection between the echo signals; only the characteristic value is used as a measurement indicator for clustering, and the change relationship and geographical connection between the clustering objects are not fully considered, resulting in deviations in the clustering results of the echo signal, which reduces the accuracy of the reinforcement quality evaluation of the rock and soil slope. Summary of the invention

[0004] In order to solve the technical problem that the clustering results of echo signals are biased due to failure to consider the change relationship and geographical connection between the echo signals of the locations to be tested, and the accuracy of the reinforcement quality evaluation of the rock and soil slope is reduced, the purpose of the present invention is to provide a data-driven intelligent evaluation method and system for the reinforcement quality of the rock and soil slope. The technical scheme adopted is as follows:

[0005] In a first aspect, an embodiment of the present invention provides a data-driven intelligent evaluation method for geotechnical slope reinforcement quality, the method comprising:

[0006] Obtaining an ultrasonic transmission signal and an ultrasonic echo signal of each to-be-tested location of the rock and soil slope, as well as a geographical location corresponding to each ultrasonic echo signal;

[0007] Combining the similarity between the ultrasonic emission signal and the ultrasonic echo signal at each location to be tested, and the distance between the maximum value points of the ultrasonic emission signal and the ultrasonic echo signal, the abnormal characteristic value of the ultrasonic echo signal at each location to be tested is obtained;

[0008] According to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal characteristic values, an initial clustering metric distance between any two ultrasonic echo signals is obtained; according to the distance between the geographical locations corresponding to any two ultrasonic echo signals, the initial clustering metric distance is adjusted to obtain a modified clustering metric distance between any two ultrasonic echo signals;

[0009] Clustering the ultrasonic echo signals based on the modified clustering metric distance to obtain different signal clustering clusters; obtaining the comprehensive abnormality degree of each signal clustering cluster according to the abnormal characteristic value of the ultrasonic echo signals in each signal clustering cluster;

[0010] The reinforcement quality of the rock and soil slope is evaluated based on the comprehensive anomaly degree.

[0011] Furthermore, the method for obtaining the abnormal characteristic value of the ultrasonic echo signal at each location to be tested by combining the similarity between the ultrasonic emission signal and the ultrasonic echo signal at each location to be tested, and the distance between the maximum value points of the ultrasonic emission signal and the ultrasonic echo signal, comprises:

[0012] The DTW algorithm is used to obtain the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be tested;

[0013] The ultrasonic transmission signal and the ultrasonic echo signal of each location to be measured are used as ultrasonic analysis signals. For each ultrasonic analysis signal, the mean value of the amplitude of all data points of the ultrasonic analysis signal is calculated as the mean energy value of the ultrasonic analysis signal.

[0014] The time between each maximum point of the ultrasonic analysis signal and its adjacent next maximum point is used as the characteristic frequency value of each maximum point of the ultrasonic analysis signal; the average value of the characteristic frequency values ​​of the remaining maximum points of the ultrasonic analysis signal except the last maximum point is calculated as the average frequency value of the ultrasonic analysis signal;

[0015] Combined with the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be tested, as well as the difference between the mean energy values ​​and the difference between the mean frequency values ​​of the ultrasonic transmission signal and the ultrasonic echo signal, the abnormal characteristic value of the ultrasonic echo signal at each location to be tested is obtained.

[0016] Furthermore, the calculation formula of the abnormal characteristic value of the ultrasonic echo signal of each location to be tested is as follows:

[0017]

[0018] Where W is the abnormal characteristic value of the ultrasonic echo signal at each location to be tested; Transmitting an ultrasonic signal for each location to be measured; An ultrasonic echo signal for each location to be tested; The DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be measured; The average energy value of the ultrasonic transmission signal at each location to be measured; The mean energy value of the ultrasonic echo signal at each location to be measured; The average frequency value of the ultrasonic transmission signal at each location to be measured; The mean frequency value of the ultrasonic echo signal of each location to be measured; is the absolute value function; Norm is the normalization function.

[0019] Furthermore, the method for obtaining the initial clustering metric distance between any two ultrasonic echo signals according to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal feature values ​​comprises:

[0020] Select any two ultrasonic echo signals as target echo signals, arrange the amplitude of the data points of each target echo signal in time sequence, and obtain the initial amplitude sequence of each target echo signal; count the total number of elements in the initial amplitude sequence of each target echo signal as the length value of the initial amplitude sequence of each target echo signal;

[0021] Determine whether the length values ​​of the two target echo signals are equal, and if so, use the initial amplitude sequence of each target echo signal as the final amplitude sequence of each target echo signal;

[0022] If not, use the target echo signal corresponding to the largest length value as the first echo signal, and use the target echo signal corresponding to the smallest length value as the second echo signal; use the initial amplitude sequence of the first echo signal as the final amplitude sequence of the first echo signal; add an element with a constant value of 0 to the end of the initial amplitude sequence of the second echo signal until the length value of the updated initial amplitude sequence of the second echo signal is equal to the length value of the initial amplitude sequence of the first echo signal, then stop updating the initial amplitude sequence of the second echo signal, and use the updated initial amplitude sequence of the second echo signal as the final amplitude sequence of the second echo signal;

[0023] An initial clustering metric distance between the two target echo signals is obtained according to the difference between the values ​​of the elements with the same subscript in the final amplitude sequences of the two target echo signals and the difference between the abnormal feature values.

[0024] Furthermore, the calculation formula of the initial clustering metric distance between the two target echo signals is as follows:

[0025]

[0026] In the formula, Measuring a distance between the first target echo signal and the second target echo signal for the initial clustering; is the abnormal characteristic value of the first target echo signal; is the abnormal characteristic value of the second target echo signal; I is the total number of elements in the final amplitude sequence of the first target echo signal; is the value of the i-th element in the final amplitude sequence of the first target echo signal; is the value of the i-th element in the final amplitude sequence of the second target echo signal; is the absolute value function.

[0027] Furthermore, the method of adjusting the initial cluster metric distance according to the distance between the geographical locations corresponding to any two ultrasonic echo signals to obtain a modified cluster metric distance between any two ultrasonic echo signals includes:

[0028] Obtaining the distance between the geographical locations corresponding to any two ultrasonic echo signals as the characteristic distance between any two ultrasonic echo signals; and obtaining the characteristic comprehensive distance based on the characteristic distance between each ultrasonic echo signal;

[0029] The ratio of the characteristic distance between any two ultrasonic echo signals to the characteristic comprehensive distance is used as the characteristic weight between any two ultrasonic echo signals; the product of the initial clustering metric distance between any two ultrasonic echo signals and the characteristic weight is normalized to obtain the modified clustering metric distance between any two ultrasonic echo signals.

[0030] Furthermore, the calculation formula of the feature comprehensive distance is as follows:

[0031]

[0032] Wherein, U is the feature comprehensive distance; is the characteristic distance between the mth ultrasonic echo signal and the nth ultrasonic echo signal; W is the total number of ultrasonic echo signals.

[0033] Furthermore, the method for obtaining the comprehensive abnormality of each signal cluster includes:

[0034] The mean value of the abnormal characteristic values ​​of all ultrasonic echo signals in each signal cluster is calculated as the initial abnormality degree of each signal cluster; and the initial abnormality degree is normalized to obtain the comprehensive abnormality degree of each signal cluster.

[0035] Furthermore, the method for evaluating the reinforcement quality of the rock and soil slope based on the comprehensive anomaly degree includes:

[0036] For each signal clustering cluster, the signal clustering cluster whose comprehensive abnormality is greater than the preset abnormality threshold is regarded as a qualified clustering cluster, and the signal clustering cluster whose comprehensive abnormality is less than or equal to the preset abnormality threshold is regarded as an abnormal clustering cluster;

[0037] The reinforcement quality of the rock and soil slope at the location to be tested corresponding to the ultrasonic echo signals in the qualified clusters is qualified, and the reinforcement quality of the rock and soil slope at the location to be tested corresponding to the ultrasonic echo signals in the abnormal clusters is unqualified.

[0038] In the second aspect, another embodiment of the present invention provides a data-driven intelligent evaluation system for rock and soil slope reinforcement quality, the system comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of any one of the above methods are implemented.

[0039] The present invention has the following beneficial effects:

[0040] In the embodiment of the present invention, the similarity between the ultrasonic emission signal and the ultrasonic echo signal at the location to be tested reflects the reinforcement condition of the rock and soil slope at the location to be tested, and the frequency characteristics of the signal, that is, the distance between the maximum points of the ultrasonic emission signal and the ultrasonic echo signal, are comprehensively analyzed so that the acquired abnormal characteristic value accurately reflects the abnormal situation of the slope reinforcement; the ultrasonic echo signals are affected by the geographical location and the spatial distance, and the abnormal characteristic value of the ultrasonic echo signal and the amplitude of the data point in the signal are further combined to obtain the initial clustering metric distance; since the geographical environment and soil moisture conditions of the locations to be tested that are close in location are relatively similar, the locations to be tested that are close in location are relatively close in location. The slope reinforcement quality of the measured locations is similar. The initial clustering measurement distance only considers the abnormal situation of the measured location presented by the ultrasonic echo signal. The present application adds the distance between the geographical locations corresponding to the ultrasonic echo signals for correction, so that the obtained corrected clustering measurement distance takes into account the influence on the spatial distance, thereby being more accurate, and avoiding errors in subsequent clustering results due to the particularity of the geographical location; the ultrasonic echo signals are clustered based on the corrected clustering measurement distance to obtain the resulting clustering clusters, so as to obtain more accurate clustering results; the reinforcement quality of the rock and soil slope is further evaluated based on the comprehensive abnormality of the signal clustering clusters obtained by analysis, so as to improve the accuracy of the reinforcement quality evaluation of the rock and soil slope. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A flowchart of the steps of a data-driven intelligent evaluation method for geotechnical slope reinforcement quality provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the data-driven intelligent evaluation method and system for rock and soil slope reinforcement quality proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0045] The following is a detailed description of a data-driven intelligent evaluation method and system for rock and soil slope reinforcement quality provided by the present invention in conjunction with the accompanying drawings.

[0046] See also Figure 1 , which shows a flow chart of steps of a data-driven intelligent evaluation method for geotechnical slope reinforcement quality provided by an embodiment of the present invention, the method comprising:

[0047] Step S1: Acquire the ultrasonic transmission signal and ultrasonic echo signal of each to-be-tested location of the rock and soil slope, as well as the geographical location corresponding to each ultrasonic echo signal.

[0048] Specifically, the present application uses ultrasonic technology to detect the reinforcement quality of rock and soil slopes. The specific operation is: an ultrasonic sensor arranged inside a drone is used to transmit an ultrasonic signal to the test location at a certain height above the test location of the rock and soil slope. The ultrasonic wave is reflected by the test location of the rock and soil slope to generate an echo, and the detector carried by the drone is used to measure the echo signal absorbed and reflected by the test location.

[0049] The ultrasonic wave emitted to the location to be tested is used as the ultrasonic emission signal of the location to be tested, and the ultrasonic wave generates an echo signal after being reflected by the location to be tested as the ultrasonic echo signal of the location to be tested; each location to be tested has an ultrasonic emission signal and an ultrasonic echo signal. At the same time, when the drone emits ultrasonic waves to the location to be tested in the air, the geographical location of the drone is used as the geographical location corresponding to the ultrasonic echo signal corresponding to the location to be tested. Afterwards, the ultrasonic emission signal and the ultrasonic echo signal of the location to be tested, as well as the geographical location corresponding to the ultrasonic echo signal, are transmitted to the remote analysis system through the wireless transmitting device arranged inside the drone for subsequent analysis. Among them, the geographical location is a well-known technology and will not be described in detail here.

[0050] It should be noted that in order to ensure the accuracy of the subsequent reinforcement quality evaluation, at least two locations to be measured are selected for measurement. The horizontal axis of the ultrasonic emission signal represents the propagation time of the sound wave, and the vertical axis represents the amplitude of the sound wave signal; the horizontal axis of the ultrasonic echo signal represents the round-trip time of the sound wave, represents the distance or depth detected by the ultrasound, and the vertical axis represents the amplitude or intensity of the echo signal.

[0051] Step S2: Based on the similarity between the ultrasonic emission signal and the ultrasonic echo signal at each location to be tested, and the distance between the maximum points of the ultrasonic emission signal and the ultrasonic echo signal, the abnormal characteristic value of the ultrasonic echo signal at each location to be tested is obtained.

[0052] Specifically, when using ultrasound to inspect the quality of the reinforcement of rock and soil slopes, if the rock and soil are uniform and have no cracks, the ultrasonic wave will not encounter obvious medium changes during the propagation process, and the similarity between the ultrasonic echo signal and the ultrasonic transmission signal is high; if there are inhomogeneities or cracks in the soil in the rock and soil slope, it is easy to cause ultrasonic scattering, etc., causing the echo signal to change, resulting in a low similarity between the ultrasonic echo signal and the ultrasonic transmission signal. The distance between the maximum point of the ultrasonic transmission signal and the ultrasonic echo signal reflects the frequency characteristics of the signal. Therefore, combined with the similarity between the ultrasonic transmission signal and the ultrasonic echo signal at the location to be tested, and the distance between the maximum point of the ultrasonic transmission signal and the ultrasonic echo signal, the obtained abnormal characteristic value is analyzed to make it more accurate.

[0053] Preferably, the specific method for obtaining the abnormal characteristic value is: using the DTW algorithm to obtain the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal of each location to be tested; using the ultrasonic transmission signal and the ultrasonic echo signal of each location to be tested as ultrasonic analysis signals, and for each ultrasonic analysis signal, calculating the mean of the amplitudes of all data points of the ultrasonic analysis signal as the mean energy value of the ultrasonic analysis signal; using the time between each maximum point of the ultrasonic analysis signal and its next adjacent maximum point as the characteristic frequency value of each maximum point of the ultrasonic analysis signal; calculating the mean of the characteristic frequency values ​​of the remaining maximum points of the ultrasonic analysis signal except the last maximum point as the mean frequency value of the ultrasonic analysis signal; combining the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal of each location to be tested, as well as the difference between the mean energy values ​​and the difference between the mean frequency values ​​of the ultrasonic transmission signal and the ultrasonic echo signal, to obtain the abnormal characteristic value of the ultrasonic echo signal of each location to be tested.

[0054] It should be noted that both the ultrasonic transmission signal and the ultrasonic echo signal are ultrasonic analysis signals. The amplitudes of the data points of the ultrasonic analysis signal are arranged in time series to obtain the initial amplitude sequence of each ultrasonic analysis signal. The DTW algorithm is used to obtain the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at the test location, which is equivalent to using the DTW algorithm to calculate the DTW value between the ultrasonic transmission signal and the initial amplitude sequence of the ultrasonic echo signal at the test location.

[0055] The calculation formula for the abnormal characteristic value of the ultrasonic echo signal at each location to be tested is as follows:

[0056]

[0057] Where W is the abnormal characteristic value of the ultrasonic echo signal at each location to be tested; Transmitting an ultrasonic signal for each location to be measured; An ultrasonic echo signal for each location to be tested; is the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be tested; is the average energy value of the ultrasonic transmission signal at each location to be tested; is the average energy value of the ultrasonic echo signal at each location to be measured; is the average frequency value of the ultrasonic transmission signal at each location to be tested; is the average frequency value of the ultrasonic echo signal at each location to be measured; is the absolute value function; Norm is the normalization function.

[0058] when When is larger, the similarity between the ultrasonic emission signal and the ultrasonic echo signal at the test site is lower, indicating that the possibility of soil inhomogeneity or cracks in the rock slope at the test site is greater, and the abnormal characteristic value W is larger; when The smaller the value, the higher the similarity between the ultrasonic emission signal and the ultrasonic echo signal at the test location, which means that the rock slope at the test location is more likely to be uniform and crack-free, and the smaller the abnormal characteristic value W is.

[0059] The amplitude of the signal reflects the energy of the signal. When the value is larger, the difference between the energy intensity of the ultrasonic transmission signal and the ultrasonic echo signal at the test site, that is, the error energy, is larger, indicating that the similarity between the ultrasonic echo signal and the ultrasonic transmission signal is lower, and the possibility of soil inhomogeneity or cracks in the rock slope at the test site is greater, and the abnormal characteristic value W is larger. The spacing between the maximum value points of the signal represents the frequency characteristics of the signal. When The larger it is, the greater the difference in frequency between the ultrasonic echo signal and the ultrasonic transmission signal at the test location, which means that the similarity between the ultrasonic echo signal and the ultrasonic transmission signal is lower, and the greater the possibility that there is soil inhomogeneity or cracks in the rock slope at the test location, the larger the abnormal characteristic value W.

[0060] Step S3: According to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal characteristic values, the initial clustering metric distance between any two ultrasonic echo signals is obtained; according to the distance between the geographical locations corresponding to any two ultrasonic echo signals, the initial clustering metric distance is adjusted to obtain the modified clustering metric distance between any two ultrasonic echo signals.

[0061] Specifically, the abnormal characteristic value of each ultrasonic echo signal is obtained through step S2. When clustering the ultrasonic echo signals, the traditional method only measures the abnormal characteristic value of the ultrasonic echo signal, that is, ultrasonic echo signals with similar abnormal characteristic values ​​may be clustered into a cluster, and the differences between the changing laws of the ultrasonic echo signals are not considered, which leads to deviations in the clustering results.

[0062] The amplitude of each data point of the ultrasonic echo signal presents the change law of the signal. The difference between the amplitudes of any two data points of the ultrasonic echo signal reflects the difference between the change laws of the ultrasonic echo signal. Combined with the difference between the abnormal feature values, the initial clustering metric distance for measuring the clustering conditions is more accurate.

[0063] Preferably, the specific method for obtaining the initial clustering metric distance is: select any two ultrasonic echo signals as target echo signals, arrange the amplitudes of the data points of each target echo signal in time series, and obtain the initial amplitude sequence of each target echo signal; count the total number of elements in the initial amplitude sequence of each target echo signal as the length value of the initial amplitude sequence of each target echo signal; determine whether the length values ​​of the two target echo signals are equal, and if so, use the initial amplitude sequence of each target echo signal as the final amplitude sequence of each target echo signal; if not, use the target echo signal corresponding to the largest length value as the first echo signal, and the target echo signal corresponding to the smallest length value as the second echo signal; taking the initial amplitude sequence of the first echo signal as the final amplitude sequence of the first echo signal; adding an element with a constant value of 0 to the end of the initial amplitude sequence of the second echo signal until the length value of the updated initial amplitude sequence of the second echo signal is equal to the length value of the initial amplitude sequence of the first echo signal, stopping updating the initial amplitude sequence of the second echo signal, and taking the updated initial amplitude sequence of the second echo signal as the final amplitude sequence of the second echo signal; obtaining an initial clustering metric distance between the two target echo signals based on the difference between the values ​​of the elements with the same subscript in the final amplitude sequences of the two target echo signals and the difference between the abnormal feature values.

[0064] It should be noted that if the lengths of the two target echo signals are the same, the total number of data points in the two target echo signals is equal, and the data points of the two target echo signals correspond one to one, for example, the first data point in the first target echo signal corresponds to the first data point in the second target echo signal, the second data point in the first target echo signal corresponds to the second data point in the second target echo signal, and so on. If the lengths of the two target echo signals are inconsistent, the data points of the two target echo signals cannot correspond one to one, and the initial amplitude sequence of the target echo signals is obtained. The embodiment of the present invention uses the maximum length value as a reference, adds elements to the end of the initial amplitude sequence of the minimum length value, and makes the total number of elements in the final amplitude sequence of the two target echo signals equal. In order to avoid the added elements affecting the accuracy of the analysis, the value of the element is set equal to the constant 0.

[0065] The calculation formula of the initial clustering metric distance between two target echo signals is as follows:

[0066]

[0067] In the formula, An initial clustering metric distance between the first target echo signal and the second target echo signal; is the abnormal characteristic value of the first target echo signal; is the abnormal characteristic value of the second target echo signal; I is the total number of elements in the final amplitude sequence of the first target echo signal; is the value of the i-th element in the final amplitude sequence of the first target echo signal; is the value of the i-th element in the final amplitude sequence of the second target echo signal; is the absolute value function.

[0068] when When the larger the value is, it means that there is a big difference in the reinforcement quality of the rock and soil slope at the test site corresponding to the two target echo signals. The larger the value, the lower the possibility that two target echo signals will be clustered into the same cluster during clustering. The difference between the changing patterns of the two target echo signals is presented. The larger the value, the greater the difference between the changing rules of the two target echo signals, indicating that there is a large difference between the two target echo signals. Therefore, the greater the difference in the soil conditions of the rock and soil slopes at the test site corresponding to the two target echo signals, the greater the initial clustering metric distance. The larger the value is, the lower the possibility that two target echo signals will be clustered into the same cluster during clustering.

[0069] The geographical environment and soil moisture of the test sites with similar locations are relatively similar. When clustering ultrasonic echo signals, the initial clustering metric distance between the ultrasonic echo signals corresponding to the test sites with similar locations is small. However, since the geotechnical conditions of the test sites with similar locations are relatively similar, although the ultrasonic echo signal with a smaller abnormal characteristic value corresponds to a relatively small abnormal degree of the test site, if the abnormal characteristic value of the ultrasonic echo signal at its neighboring location is large, due to the particularity of the geographical location, the ultrasonic echo signal with a smaller abnormal characteristic value may have potential risks; when evaluating the geotechnical slope reinforcement at this location, the ultrasonic echo signal with a smaller abnormal characteristic value needs to be clustered into the cluster with a larger abnormal characteristic value adjacent to its location. Therefore, it is necessary to further analyze the distance between the geographical locations corresponding to the ultrasonic echo signals, so as to correct the initial clustering metric distance and obtain the corrected clustering metric distance.

[0070] Preferably, the specific method for obtaining the modified clustering metric distance is: obtaining the distance between the geographical locations corresponding to any two ultrasonic echo signals as the characteristic distance between any two ultrasonic echo signals; obtaining the characteristic comprehensive distance based on the characteristic distance between each ultrasonic echo signal; taking the ratio of the characteristic distance between any two ultrasonic echo signals to the characteristic comprehensive distance as the characteristic weight between any two ultrasonic echo signals; normalizing the product of the initial clustering metric distance between any two ultrasonic echo signals and the characteristic weight to obtain the modified clustering metric distance between any two ultrasonic echo signals.

[0071] It should be noted that the distance between two geographical locations can be calculated by spherical trigonometry, which is a well-known technique and will not be described in detail here.

[0072] The calculation formula of the modified clustering metric distance between any two ultrasonic echo signals is as follows:

[0073]

[0074]

[0075] In the formula, is the modified clustering metric distance between the a-th ultrasonic echo signal and the b-th ultrasonic echo signal; is the initial clustering metric distance between the a-th ultrasonic echo signal and the b-th ultrasonic echo signal; is the characteristic distance between the ath ultrasonic echo signal and the bth ultrasonic echo signal; U is the characteristic comprehensive distance; is the feature weight between the a-th ultrasonic echo signal and the b-th ultrasonic echo signal; is the characteristic distance between the mth ultrasonic echo signal and the nth ultrasonic echo signal; W is the total number of ultrasonic echo signals, which is equal to the total number of locations to be measured; Norm is the normalization function.

[0076] Equivalent to using right Normalization processing is performed. In other embodiments of the present invention, the characteristic distance between two ultrasonic echo signals can be calculated, and the maximum characteristic distance is used as the characteristic comprehensive distance U. Since the geographical environment and soil moisture conditions of the test sites with similar locations are relatively similar, the slope reinforcement quality of the test sites with similar locations is similar; when The smaller the distance, the closer the location of the test site corresponding to the ath ultrasonic echo signal and the bth ultrasonic echo signal. The closer the slope reinforcement quality of the test site corresponding to the two ultrasonic echo signals is, the greater the possibility that the two ultrasonic echo signals will be clustered into the same cluster when clustering. That is, the metric distance between the two ultrasonic echo signals when clustering is the corrected clustering metric distance. It should be smaller to avoid errors in subsequent clustering results due to the particularity of the geographical location.

[0077] Step S4: clustering the ultrasonic echo signals based on the modified clustering metric distance to obtain different signal clustering clusters; obtaining the comprehensive abnormality degree of each signal clustering cluster according to the abnormal characteristic value of the ultrasonic echo signals in each signal clustering cluster.

[0078] In the embodiment of the present invention, the K-means clustering algorithm is used to cluster the ultrasonic echo signals corresponding to all the locations to be tested based on the modified clustering metric distance between the ultrasonic echo signals to obtain different signal clustering clusters. The number of signal clusters, i.e., K, is obtained by the elbow method. The K-means clustering algorithm and the elbow method are both well-known technologies and will not be described in detail here.

[0079] The specific method for obtaining the comprehensive abnormality of the signal clustering cluster is: calculate the mean of the abnormal characteristic values ​​of all ultrasonic echo signals in each signal clustering cluster as the initial abnormality of each signal clustering cluster; normalize the initial abnormality to obtain the comprehensive abnormality of each signal clustering cluster. The calculation formula of the comprehensive abnormality of each signal clustering cluster is as follows:

[0080]

[0081] Where, Q is the comprehensive abnormality of each signal cluster; J is the total number of ultrasonic echo signals in each signal cluster; is the abnormal characteristic value of the jth ultrasonic echo signal in each signal cluster; is the initial abnormality of each signal cluster; Norm is the normalization function.

[0082] The larger the comprehensive anomaly degree Q is, the greater the possibility that there is an abnormality in the reinforcement quality of the rock and soil slope at the test location corresponding to the ultrasonic echo signal in the signal cluster.

[0083] Step S5: Evaluate the reinforcement quality of the geotechnical slope based on the comprehensive anomaly degree.

[0084] For each signal clustering cluster, the signal clustering cluster with a comprehensive abnormality greater than the preset abnormality threshold is regarded as a qualified clustering cluster, and the signal clustering cluster with a comprehensive abnormality less than or equal to the preset abnormality threshold is regarded as an abnormal clustering cluster; the reinforcement quality of the rock and soil slope at the test location corresponding to the ultrasonic echo signal in the qualified clustering cluster is qualified, and the reinforcement quality of the rock and soil slope at the test location corresponding to the ultrasonic echo signal in the abnormal clustering cluster is unqualified.

[0085] It should be noted that the preset abnormal threshold is an empirical value of 0.3, and implementers can set it according to specific circumstances.

[0086] So far, the present invention is completed.

[0087] In summary, in an embodiment of the present invention, the abnormal characteristic value of the ultrasonic echo signal of the location to be tested is obtained, and the initial clustering metric distance is obtained according to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal characteristic values; the initial clustering metric distance is adjusted according to the distance between the geographical locations corresponding to any two ultrasonic echo signals to obtain a modified clustering metric distance; the ultrasonic echo signals are clustered based on the modified clustering metric distance to obtain signal clustering clusters; the reinforcement quality of the rock and soil slope is evaluated based on the comprehensive abnormality of the obtained signal clustering clusters. The present invention takes into account the spatial distance between ultrasonic echo signals, improves the accuracy of the clustering results based on the modified clustering metric distance, and improves the accuracy of the reinforcement quality evaluation of the rock and soil slope.

[0088] Based on the same inventive concept as the above method embodiment, the embodiment of the present invention further provides a data-driven intelligent evaluation system for geotechnical slope reinforcement quality, the system comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above data-driven intelligent evaluation method for geotechnical slope reinforcement quality are implemented, such as Figure 1 The data-driven intelligent evaluation method for rock slope reinforcement quality has been described in detail in the above embodiment and will not be repeated here.

[0089] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0090] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A data-driven intelligent evaluation method for geotechnical slope reinforcement quality, characterized in that: The method includes: Obtaining an ultrasonic transmission signal and an ultrasonic echo signal of each to-be-tested location of the rock and soil slope, as well as a geographical location corresponding to each ultrasonic echo signal; Combining the similarity between the ultrasonic emission signal and the ultrasonic echo signal at each location to be tested, and the distance between the maximum value points of the ultrasonic emission signal and the ultrasonic echo signal, the abnormal characteristic value of the ultrasonic echo signal at each location to be tested is obtained; According to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal characteristic values, an initial clustering metric distance between any two ultrasonic echo signals is obtained; according to the distance between the geographical locations corresponding to any two ultrasonic echo signals, the initial clustering metric distance is adjusted to obtain a modified clustering metric distance between any two ultrasonic echo signals; Clustering the ultrasonic echo signals based on the modified clustering metric distance to obtain different signal clustering clusters; obtaining the comprehensive abnormality degree of each signal clustering cluster according to the abnormal characteristic value of the ultrasonic echo signals in each signal clustering cluster; evaluating the reinforcement quality of the rock and soil slope based on the comprehensive anomaly degree; The method for obtaining the abnormal characteristic value of the ultrasonic echo signal at each location to be tested by combining the similarity between the ultrasonic emission signal and the ultrasonic echo signal at each location to be tested, and the distance between the maximum value points of the ultrasonic emission signal and the ultrasonic echo signal, comprises: The DTW algorithm is used to obtain the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be tested; The ultrasonic transmission signal and the ultrasonic echo signal of each location to be measured are used as ultrasonic analysis signals. For each ultrasonic analysis signal, the mean value of the amplitude of all data points of the ultrasonic analysis signal is calculated as the mean energy value of the ultrasonic analysis signal. The time between each maximum point of the ultrasonic analysis signal and its adjacent next maximum point is used as the characteristic frequency value of each maximum point of the ultrasonic analysis signal; the average value of the characteristic frequency values ​​of the remaining maximum points of the ultrasonic analysis signal except the last maximum point is calculated as the average frequency value of the ultrasonic analysis signal; Acquire the abnormal characteristic value of the ultrasonic echo signal at each location to be tested by combining the DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be tested, and the difference between the mean energy values ​​and the difference between the mean frequency values ​​between the ultrasonic transmission signal and the ultrasonic echo signal; The method for obtaining the initial clustering metric distance between any two ultrasonic echo signals according to the difference between the amplitudes of the data points of any two ultrasonic echo signals and the difference between the abnormal feature values ​​comprises: Select any two ultrasonic echo signals as target echo signals, arrange the amplitude of the data points of each target echo signal in time sequence, and obtain the initial amplitude sequence of each target echo signal; count the total number of elements in the initial amplitude sequence of each target echo signal as the length value of the initial amplitude sequence of each target echo signal; Determine whether the length values ​​of the two target echo signals are equal, and if so, use the initial amplitude sequence of each target echo signal as the final amplitude sequence of each target echo signal; If not, use the target echo signal corresponding to the largest length value as the first echo signal, and use the target echo signal corresponding to the smallest length value as the second echo signal; use the initial amplitude sequence of the first echo signal as the final amplitude sequence of the first echo signal; add an element with a constant value of 0 to the end of the initial amplitude sequence of the second echo signal until the length value of the updated initial amplitude sequence of the second echo signal is equal to the length value of the initial amplitude sequence of the first echo signal, then stop updating the initial amplitude sequence of the second echo signal, and use the updated initial amplitude sequence of the second echo signal as the final amplitude sequence of the second echo signal; Acquire an initial clustering metric distance between the two target echo signals according to the difference between the values ​​of the elements with the same subscript in the final amplitude sequences of the two target echo signals and the difference between the abnormal feature values; The method of adjusting the initial cluster metric distance according to the distance between the geographical locations corresponding to any two ultrasonic echo signals to obtain a modified cluster metric distance between any two ultrasonic echo signals comprises: Obtaining the distance between the geographical locations corresponding to any two ultrasonic echo signals as the characteristic distance between any two ultrasonic echo signals; and obtaining the characteristic comprehensive distance based on the characteristic distance between each ultrasonic echo signal; The ratio of the characteristic distance between any two ultrasonic echo signals to the characteristic comprehensive distance is used as the characteristic weight between any two ultrasonic echo signals; the product of the initial clustering metric distance between any two ultrasonic echo signals and the characteristic weight is normalized to obtain the modified clustering metric distance between any two ultrasonic echo signals; The calculation formula of the feature comprehensive distance is as follows: Wherein, U is the feature comprehensive distance; is the characteristic distance between the mth ultrasonic echo signal and the nth ultrasonic echo signal; W is the total number of ultrasonic echo signals.

2. According to claim 1, a data-driven intelligent evaluation method for rock and soil slope reinforcement quality is characterized in that: The calculation formula of the abnormal characteristic value of the ultrasonic echo signal of each location to be tested is as follows: Where W is the abnormal characteristic value of the ultrasonic echo signal at each location to be tested; Transmitting an ultrasonic signal for each location to be measured; An ultrasonic echo signal for each location to be tested; The DTW value between the ultrasonic transmission signal and the ultrasonic echo signal at each location to be measured; The average energy value of the ultrasonic transmission signal at each location to be measured; The mean energy value of the ultrasonic echo signal at each location to be measured; The average frequency value of the ultrasonic transmission signal at each location to be measured; The mean frequency value of the ultrasonic echo signal of each location to be measured; is the absolute value function; Norm is the normalization function.

3. The data-driven intelligent evaluation method for rock and soil slope reinforcement quality according to claim 1 is characterized in that: The calculation formula of the initial clustering metric distance between the two target echo signals is as follows: In the formula, Measuring a distance between the first target echo signal and the second target echo signal for the initial clustering; is the abnormal characteristic value of the first target echo signal; is the abnormal characteristic value of the second target echo signal; I is the total number of elements in the final amplitude sequence of the first target echo signal; is the value of the i-th element in the final amplitude sequence of the first target echo signal; is the value of the i-th element in the final amplitude sequence of the second target echo signal; is the absolute value function.

4. The data-driven intelligent evaluation method for rock and soil slope reinforcement quality according to claim 1 is characterized in that: The method for obtaining the comprehensive abnormality of each signal cluster comprises: The mean value of the abnormal characteristic values ​​of all ultrasonic echo signals in each signal cluster is calculated as the initial abnormality degree of each signal cluster; and the initial abnormality degree is normalized to obtain the comprehensive abnormality degree of each signal cluster.

5. The data-driven intelligent evaluation method for geotechnical slope reinforcement quality according to claim 1 is characterized in that: The method for evaluating the reinforcement quality of a rock and soil slope based on the comprehensive anomaly degree comprises: For each signal clustering cluster, the signal clustering cluster whose comprehensive abnormality is greater than the preset abnormality threshold is regarded as a qualified clustering cluster, and the signal clustering cluster whose comprehensive abnormality is less than or equal to the preset abnormality threshold is regarded as an abnormal clustering cluster; The reinforcement quality of the rock and soil slope at the location to be tested corresponding to the ultrasonic echo signals in the qualified clusters is qualified, and the reinforcement quality of the rock and soil slope at the location to be tested corresponding to the ultrasonic echo signals in the abnormal clusters is unqualified.

6. A data-driven intelligent evaluation system for geotechnical slope reinforcement quality, characterized in that: The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a data-driven intelligent evaluation method for rock and soil slope reinforcement quality as described in any one of claims 1 to 5 are implemented.